JP2002323369A - Operation method of machine, inference method of vibration and system thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for detecting vibration by using amplitude modulation or frequency modulation of a pulse train. SOLUTION: A machine such as a gas turbine 203 includes a rotor 208 for supporting a gear 4. A sensor 8 positioned nearby generates the pulse train 72 with passage of a tooth 16 of the gear 4. The frequency of the pulse train 72 shows the rotational speed of the rotor. In addition, vibration of the rotor 208 allows the gear 4 to perform orbital motion with another center. The orbital motion generates the amplitude modulation or the frequency modulation of the pulse train 72, or both modulations. Detection of the modulation indicates existence of the vibration. In this way, both the speed and existence of the vibration are indicated by using one pulse train 72 generated by one sensor 4.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to vibration sensing.

[0002]

2. Description of the Related Art Gas turbine engines typically include one or more accelerometers for detecting vibration. Since the accelerometer may malfunction, a backup accelerometer is often provided. The use of accelerometers increases the weight of the engine. Also, the costs for manufacturing, design and maintenance increase. In addition, some accelerometers are fragile and easily damaged.

[0003]

SUMMARY OF THE INVENTION The present invention alleviates some or all of the disadvantages identified above. One form of the present invention detects vibrations by analyzing existing pulse trains generated by existing sensors and currently used in velocity measurement applications. According to the present invention, both speed and vibration are indicated using an existing pulse train.

[0004]

DETAILED DESCRIPTION OF THE INVENTION The present invention utilizes a pulse train generated by a sensor. Many types of sensors can be used, but for simplicity, a general reluctance sensor is described below.

FIG. 1 shows a gear 4 and a reluctance sensor 8.
1 shows a conventional system including an electronic circuit 12. Electronic circuit 12 detects each tooth 16 passing through reluctance sensor 8. Electronic circuit 12 generates one pulse (not shown) from output terminal 17 in response to each tooth 16.

"Reluctance" is magnetoresistance. Generally, the magnetoresistance is determined by (1) the amount of the substance located inside the region surrounded by the broken line 20, and (2) the magnetic permeability. For example, tooth 16 in FIG. 2 is located above reference line 24.
The reluctance detected by the sensor 8 is represented by the point 28 in the characteristic diagram 38.
Are indicated by.

As a second example, the tooth 16 in FIG.
On top of 4. The reluctance at this time is indicated by a point 32. As a third example, tooth 16 in FIG. 4 is located below reference line 24. The reluctance at this time is indicated by a point 36.

FIG. 5 is a characteristic diagram 39 showing a generalized relationship between reluctance and position. The characteristic diagram is reference line 24.
And the minimum point 4 corresponding to point 32 in FIG.
Has zero.

The electronic circuit 12 does not always measure reluctance itself, but often measures parameters related to reluctance. For example, FIG. 6 shows a shaded tooth 16. The shaded tooth 16 passes through the sensor 8 and moves to the position shown in FIG. In response to this movement, the electronic circuit 12 generates a signal 41 similar to the signal shown in FIG.
Occurs. Due to the measurement technique used, the signal 41 has a slope greater than the slope of the reluctance characteristic diagram 39 of FIG. 5 as compared to the reluctance itself.

For the sake of simplicity, the characteristic diagram of FIG. 8 does not strictly show the slopes of all the points in the characteristic diagram 39 of FIG. 5, but shows only the general features thereof. In FIG. 8, the negative region 48 corresponds to the region 52 of FIG. 5 having a negative slope. The region 56 in FIG. 8 that is positive is the region 56 in FIG.
Region 60. A point 64 in FIG. 8 having a value of 0 corresponds to a point 40 in FIG. 5 having a slope of 0.

As the teeth 16 successively pass through the reluctance sensor 8 of FIG. 7, a train 72 of pulses 74 is generated as shown in FIG. When using a reluctance sensor that measures the actual reluctance instead of the slope,
The pulse train (not shown) includes a series of the characteristic diagrams 39 of FIG.

In the ideal case, the pulse train 72 of FIG.
Pulse 74 has exactly the same shape, and the time interval 76 between adjacent pulses is equal. In the ideal case, the gear 4 of FIG. 7 must be perfectly symmetric, have exactly the same permeability and rotate at a constant speed with respect to the fixed center 82 of FIG. It is.

However, when vibration occurs, this ideal state disappears. The gear 4 not only rotates about the center 82, but also causes the center 82 to orbit about another center. FIG. 10 illustrates such a situation. Disk 8
Reference numeral 6 denotes the gear 4 of FIG. The disk 86 in FIG. 10 is supported by a shaft 88 and rotates about an axis 90. The center 82 is shown.

Further, to show the orbital motion, the shaft 88
Are supported by the second disk 94. Second disk 9
4 rotates about a second axis 98. FIG.
The positions occupied by the respective components while the components perform the rotation and the orbital motion in combination with each other are shown in order.

A characteristic diagram 100 in FIG. 11 shows a relative arrangement of each component at an initial reference time. The reluctance sensor 8 and the shaft 88 are shown. Arm 107 is shown overlaid to show that disk 94 acts as a crank arm when supporting shaft 88.
Arm 107 rotates about center 98. A second arm 105 is also shown to show that the disc 86 acts as a crank arm when supporting the reference box 106 representing the teeth 16 of FIG. In FIG. 11, the arm 105 rotates around a shaft 88.

For simplicity, the two disks 86 and 94
And the two crank arms 105 and 107 rotate at the same angular velocity.

FIG. 11 shows seven characteristic diagrams. Table 1 below shows the amount of rotation that occurs in each plot. As described above, the two crank arms 105 and 107 have different centers, but rotate at the same angular velocity. Therefore, Table 1 shows one rotation amount for each characteristic diagram.

Since the motor rotates at the same speed at any time, the angular displacement from the initial position in the characteristic diagram 100 becomes the same. That is, at any time, the crank arm 1
Both 05 and 107 have different centers of rotation, but rotate by the same total amount of rotation.

[0019]

[Table 1]

The combination rotation of FIG. 11 has two important features. One is that the disk 8
6 and the reluctance sensor 8 change in distance. As a specific example, distance 1 in the characteristic diagram 124 is used.
28 is larger than the distance 132 in the characteristic diagram 108. The change in distance changes the reluctance signal generated by the sensor 8.

The second feature is that as the combined rotation and orbital motion occurs, the disc 86 will move to the reluctance sensor 8.
Is the speed at which it passes. FIG. 12 illustrates this change and includes the characteristic diagrams 100 and 124 of FIG. 11 as they are. All rotations are counterclockwise. For simplicity, FIG. 12 only considers the vertical velocity component. In the center of the figure, "up" and "down" are indicated.

In characteristic diagram 100, vector 140 represents the speed of shaft 88 in the vertical direction. Since the shaft 88 is the axis of rotation of the disk 86, the vector 140
Also represents the translation speed of the entire disk 86 in the upward direction. Since reference box 106 is mounted on disk 86, vector 140 also represents one velocity component of box 106 in the upward direction.

In addition, vector 144 indicates that box 86 due to rotation of disk 86 about shaft 88.
Represents the additional speed. The net velocity of box 106 in the upward direction is the vector sum of vector 140 and vector 144. The net speed is relatively fast compared to the net speed in the case of the characteristic diagram 124 considered below.

In the characteristic diagram 124, the shaft 88 is
Is moving downward to rotate counterclockwise with respect to. Vector 148 represents its downward velocity component.
Since the disk 86 is mounted on the shaft 88, the vector 148 also represents the downward translation speed of the entire disk 86.
Accordingly, box 106 will have a downward velocity component as indicated by arrow 148 due to the downward translation of disk 86.

In addition, the rotation of disk 86 about shaft 88 causes box 106 to have an upward velocity component. Vector 152 represents that component.
The net velocity of the vertical box 106 is the vector 148
And vector 152. Characteristic chart 124
Since the vector 148 and the vector 152 are opposite to each other, the net speed is relatively slower than the net speed in the case of the characteristic diagram 100.

Thus, the vibration of the disk 86 in FIG. 10 can take the form of a orbital movement about a shaft or axis 98 of the center 88. The disk 86 represents the gear 4 of FIG. This orbital movement is caused by the sensor 8 and the electronic circuit 12 shown in FIG.
Causes two events in the parameter measured by

One event is that a change occurs in the reluctance signal due to a change in reluctance sensed by the sensor 8 due to the orbital movement. The second event is that the tangential velocity when the circumference of the disk 86 in FIG. Since the teeth 16 in FIG. 6 are located around them, as the orbital motion occurs, the speed of the teeth 16 also changes.

FIGS. 13 and 15 show the pulse train 72 of FIG.
Shows the effects of these events. FIG. 13 illustrates some type of amplitude modulation. That is, the amplitude at point 160 is larger than the amplitude at point 164. This amplitude change occurs because the disk 86 moves toward and away from the sensor 8 in FIG.
Orbital movement about axis 98 causes this movement.
Amplitude is measured from points like 0 to 160.

Most of the pulses shown in FIG.
Not symmetric about 63. The reason is complicated,
This depends in part on the technique used to generate the illustrated pulse train. However, one of the factors affecting the lack of symmetry is shown in FIG.

In FIG. 14, characteristic diagrams 170, 174,
Reference numerals 175 and 179 indicate four consecutive positions of the reference block 165. These four positions are shown superimposed on the characteristic diagram 183, and the corresponding characteristic diagram numbers are given.

The characteristic diagram 183 shows that the path of the block 165 is not symmetric about the axis 24. The lack of symmetry is partly due to the lack of symmetry about the zero amplitude axis 163 as shown in FIG. For example, in a very general sense, point 160 in FIG.
May correspond to the position of the block 165 in the characteristic diagram 170 of FIG. The point 161 in FIG. 13 corresponds to the position of the block 165 in the characteristic diagram 179 in FIG. 14, and the reluctance in this case is somewhat lower. The characteristic diagram 183 of FIG. 14 shows the two positions in one characteristic diagram, and the difference in reluctance can be more clearly understood.

FIG. 15 shows some frequency modulation, where the frequency at time 184 is higher than at time 188. This increase in frequency, ie, the shortening of the time interval between adjacent pulses, will occur, for example, in the characteristic diagram 100 of FIG. In the characteristic diagram 100, the tangential speed is relatively high as described with reference to FIG.

A decrease in frequency, ie, an increase in the time interval between adjacent pulses, will occur in the characteristic diagram 124 of FIG. In the characteristic diagram 124, the tangential speed is relatively low.

Thus, two changes occur as the disk 86 rotates and orbits in FIG. One change is a change in the distance between the disk 86 and the sensor 8 in FIG. The change causes a change in reluctance. The change in reluctance causes amplitude modulation of the pulse train as shown in FIG.

The second change is a change in the tangential speed of the disk 86. The change in speed causes frequency modulation as shown in FIG.

FIG. 16 is a flow chart of the logic used to detect amplitude modulation and frequency modulation shown in FIGS. Block 190 indicates that a pulse train such as the pulse train of FIG. 9 is received. This pulse train may or may not include the amplitude modulation or frequency modulation shown in FIGS.

The block 192 in FIG. 16 corresponds to the gear 4 in FIG.
Is calculated. For example, assume that the spacing between gear teeth 16 is 10 degrees. When 15 pulses are counted in 0.01 second, the rotation speed becomes (1
Calculated as 5 × 10) degrees / 0.01 second. This quotient is equivalent to 150,000 degrees per second, or approximately 41 rpm.

Block 194 indicates that amplitude modulation has been detected. Such detection is well known and many different techniques can be used. As a simple example, the amplitude of each pulse 74 of FIG. 9 can be stored in a stack memory. The stack memory may include 1000 storage locations. When the stack fills, the earliest stored amplitude in the stack is lost.

As a specific example, stacks with amplitudes from 1 to 1
000 are sequentially stored. At this point the stack is full. When the amplitude 1001 is added, the amplitude 1 is lost. If amplitude 1002 is added, amplitude 2 will be lost, and so on.

The detection routine searches for an amplitude deviation stored in the stack. As a simple example, the detection routine may scan the stack to find both maximum and minimum amplitudes. If the difference between the maximum amplitude and the minimum amplitude exceeds a threshold, it is inferred that there is unacceptable vibration.

Block 196 of FIG. 16 indicates that a frequency modulation has been detected. Such detection is well known and many different techniques can be used. As a simple example, a second stack that stores a pair of adjacent amplitudes among the 1000 amplitudes stored in the first stack and stores a time interval of each amplitude pair may be used. The detection routine scans the second stack looking for maximum and minimum intervals. If the difference between the maximum interval and the minimum interval exceeds a threshold, it is inferred that there is unacceptable vibration.

Block 198 indicates that an alert is issued if an unacceptable vibration is found. For example,
A warning signal can be sent to the cockpit of the aircraft if either the amplitude modulation or the frequency modulation exceeds a limit.

Alternatively, numerical values indicating the amount of frequency modulation and the amount of amplitude modulation can be displayed to an operator such as a pilot. In communication services, carrier modulation is commonly expressed as a percentage, such as 50% modulation. In the present invention, this convention can be used.

It is also possible to employ other more complex methods of detecting the modulation. For example, one goal may be to detect excess deviation in frequency and amplitude of the measured pulse train from the ideal pulse train. To identify the deviation, a fast Fourier transform FFT of the pulse train is determined.

If the pulse train is an ideal pulse train containing identical pulses at equal intervals, the pulse train exhibits a predetermined distribution of the Fourier term. Furthermore, if the pulse is a true sine wave, there will be a single Fourier term.

Modulation of either the amplitude or frequency of the pulse train changes the terms of the Fourier series. If the change exceeds a threshold, it is inferred that there is unacceptable vibration.
As a simple example, infer that there is unacceptable oscillation if the fundamental frequency term and the three lowest three harmonics each change by 10 percent. More generally, if any of the first N harmonics changes by X percent each, it is inferred that there is unacceptable vibration.

FIG. 17 shows one embodiment of the present invention. High pressure compressor 200, high pressure turbine 204, fan 208
And a turbofan aircraft engine 203 including a low pressure turbine 212. Gear 4 is used to measure the speed of fan 208. The gear 4 need not actually function as a gear, but can only be used as a gear for the sole purpose of generating pulses.

Block 216 represents the reluctance sensor generating pulse train 72 of FIG. 9 and its associated electronics.

The calculations indicated by the flow chart of FIG. 16 are performed by the device represented by block 220 in FIG. Alternatively, the digital engine control device 2
24 may also perform the calculations of block 220.

The digital engine controller 224 is known in the art and measures various operating parameters such as component speed, airflow and pressure. Based on those parameters, the digital engine controller schedules or controls other parameters such as fuel / air ratio, blade cooling and stator blade angle. Digital engine controller 224 is a microprocessor (not shown) capable of performing the calculations described above in connection with FIG.
including.

Although the reluctance sensor has been described above, the reluctance sensor is not required. Other sensors can also generate the pulse train of FIG. 9 in response to the passage of a gear tooth. The sensor used should generate pulses of different magnitude when the distance to the tooth changes.
Also, the sensor should generate a pulse in response to the passage of tooth 16 in FIG. 1 so that the pulse frequency changes as the tooth acceleration changes. Some examples of sensors include Hall effect sensors, optical proximity sensors, and microwave proximity sensors.

Many substitutions and modifications can be made without departing from the true spirit of the invention. What is desired to be guaranteed by a patent is the invention as defined in the following claims.

[Brief description of the drawings]

FIG. 1 is a diagram showing a gear 4 and a conventional reluctance sensor.

FIG. 2 shows three different parts of the tooth 16 and their corresponding reluctance detected by the sensor 8;

FIG. 3 shows three different parts of a tooth 16 and their corresponding reluctance detected by a sensor 8;

FIG. 4 shows three different parts of a tooth 16 and their corresponding reluctance detected by a sensor 8;

5 is a characteristic diagram showing a relationship between the reluctance of FIG. 4 detected by a sensor 8 and an angular position.

FIG. 6 is a diagram showing a state in which one tooth 16 passes through a sensor 8;

FIG. 7 is a diagram showing a state in which one tooth 16 passes through a sensor 8;

8 shows one pulse generated by the electronic circuit 12 of FIGS. 6 and 7 when a tooth 16 passes through the sensor 8. FIG.

FIG. 9 shows a state in which a tooth repeatedly passes through the sensor 8;
FIG. 6 is a diagram showing a pulse train 72 generated by the electronic circuit 12 of FIG.

10 shows the vibration of a disk 86 representing the gear 4 of FIG.
The figure which shows how it can represent as the orbital motion of the shaft 88 centering on 8.

FIG. 11 shows several different rotational positions of the device of FIG.

FIG. 12 is used to indicate speed changes, FIG.
FIG.

FIG. 13 is a diagram showing amplitude modulation of the pulse train 72 in FIG. 9;

FIG. 14 is a diagram showing how the reference block 165 showing the teeth 16 of FIG. 1 follows the asymmetrical path when the orbital motion of the disk 86 occurs, thereby generating the amplitude modulation of FIG. 13;

FIG. 15 is a diagram showing frequency modulation of the pulse train 72 in FIG. 9;

FIG. 16 is a flowchart showing a procedure realized by one embodiment of the present invention.

FIG. 17 illustrates one embodiment of the present invention.

[Explanation of symbols]

4 gear, 8 reluctance sensor, 12 electronic circuit,
16 tooth, 72 pulse train, 82 axis, 203 turbofan aircraft engine, 208 fan, 224 digital engine controller

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Daniel Edward Mollman 11212 F-term (reference), Gosling Road, Cincinnati, Ohio, USA 2G064 AA11 AB01 BD08 CC43

Claims (20)

[Claims] 1. A method of operating a machine comprising: a) extracting a pulse train from a rotating component; b) using the pulse train; i) calculating a rotational speed of the component; and ii) the configuration. Inferring vibrations in the element. 2. The method of claim 1, wherein said machine is a gas turbine engine. 3. The method according to claim 1, wherein the rotating component is a tooth that sequentially intersects the detection area of the sensor. 4. The method of claim 3, wherein the sensor generates a train of pulses, and wherein the vibration of the component causes a modulation of the pulse, from which the vibration is inferred. 5. The method according to claim 1, wherein the vibration of the component is inferred from a modulation of a train of pulses generated by the rotation of the component. 6. a) rotating one component of the gas turbine engine, the disk including a disk having a plurality of teeth distributed around it; b) maintaining a sensor near the disk; The sensor generates i) a pulse of one kind when one tooth passes when the disc is in the first position, and ii) one pulse when the disc is in the second position. Generating a second type of pulse upon passage of the tooth; c) calculating a rotational speed of the component based on a frequency of the pulse; d) determining the configuration based on a pulse modulation. Inferring the vibration of the element. 7. The method of claim 6, wherein the vibration of the component causes an amplitude modulation of the pulse, and the vibration is inferred when the amplitude modulation exceeds a limit. 8. The method according to claim 6, wherein the vibration of the component causes a frequency modulation of the pulse, and the vibration is inferred when the frequency modulation exceeds a limit. 9. The method according to claim 6, wherein the sensor detects a change in magnetoresistance caused by passage of a tooth. 10. A gas turbine engine (203) including a rotor; b) means (216) for generating a train of signals in response to rotation of the rotor; c) based on the train of signals. Means (220) for calculating the rotational speed of the rotor; d) i) retrieving the modulation of the sequence of signals; ii) detecting when a modulation is found, generating a warning signal indicating the modulation. Means (224). 11. The system of claim 10, wherein the alert signal indicates an amount of amplitude modulation. 12. The system according to claim 10, wherein the warning signal indicates the amount of frequency modulation. 13. The system according to claim 10, wherein the warning signal indicates that the modulation has exceeded a limit. 14. The system according to claim 1, wherein said detection means comprises a system for extracting Fourier coefficients of a sequence of signals.
The system according to any one of claims 0 to 13. 15. The system according to claim 1, wherein said detecting means further comprises a system for detecting a change in Fourier coefficients.
4. The system according to 4. 16. A gas turbine engine (203) including at least one component (208) rotatable about an axis (82); and b) i) a plurality of teeth (16) along a circumference. Ii) a disk (4) mounted on said component (208); iii) a disk (4) coaxial with the axis (82); c) a sensor (8) which generates a train of signals as the train of teeth passes. D) i) detecting modulation of the sequence of signals, and ii) detecting means (216) for generating a signal when the modulation is detected. 17. The system according to claim 16, wherein said detecting means (216) detects amplitude modulation. 18. The system according to claim 16, wherein said detecting means (216) detects frequency modulation. 19. The system according to claim 16, wherein said detecting means (216) extracts a Fourier spectrum from the sequence of signals. 20. The apparatus according to claim 16, wherein said detecting means detects a change in a Fourier spectrum.
The system according to claim 1.